The goal of this project is to determine the molecular mechanisms which isolate pathways of transcriptional regulation, using the unicellular eukaryote Saccharomyces cerevisiae. We have identified two pathways of transcriptional regulation which show many parallels. The HO and CTS1 genes show identical patterns of cell cycle regulation, with expression in late G1 phase of the cell cycle. These genes are regulated by zinc-finger containing transcription factors, SWI5 and ACE2. These two transcription factors show identical patterns of cell cycle regulated transcription and cell cycle regulated nuclear localization. Furthermore, the DNA binding domains of SWI5 and ACE2 are identical, and they both bind to the HO promoter in vitro. However, despite these many similarities, SWI5 and ACE2 regulate different genes in vivo. SWI5 activates transcription of HO but not of CTS1, and ACE2 activates CTS1 but not HO. We propose to investigate the mechanisms which isolate the SWI5/HO and ACE2/CTS1 pathways. We have data supporting two distinct models for regulation, and these two models are not mutually exclusive. The first model proposes that different negative regulators bind to the HO and CTS1 promoters, and that SWI5 and ACE2 differ in their ability to overcome the repressive effects of these promoter specific negative regulators. In vivo analysis of promoter constructs provides support for this model. The second model proposes that additional factors are required for specificity in promoter recognition, and that cooperative interactions between factors contributes to promoter binding and pathway isolation. In support of this model we have identified a factor which binds cooperatively, along with SWI5, to the HO promoter. We postulate the existence of an analogous factor which is needed for specific promoter recognition of the CTS1 promoter by ACE2. Experiments are proposed to analyze the interactions of these factors with sites on DNA and with each other, and to test models of regulatory control. The factor that binds DNA cooperatively with SWI5 has been purified. The gene encoding this factor has been cloned, and we know this factor is a homeodomain protein. Cooperative interactions have not been observed before between a zinc finger protein and a homeodomain protein, and we propose to characterize their interaction. Experiments are proposed to identify regions of these proteins required for cooperative interactions.